Kinfra VL Embedding 2B API

Name: Kinfra VL Embedding 2B
Brand: Tencent (Kinfra)
Price: 0.0769 USD
Availability: InStock

Tencent Kinfra VL Embedding 2B is a fast, compact multimodal model producing vision-language embeddings across text, image and video — at provider list price, zero platform fee.

$0.0769

Input / 1M tokens

Multi

Text + image + video

Output (embeddings)

Platform fee

Why use Kinfra VL Embedding 2B on NovAI?

Multimodal vectors - embed text, images and video in one space
Fast & compact - 2B model for high-throughput pipelines
Cross-modal retrieval - search media with text queries
Object-array input - pass [{"type":"text"|"image_url","...":...}]
Zero platform fee - raw provider list price, no markup
OpenAI-compatible - via /v1/embeddings (multimodal endpoint)

Best use cases

High-volume cross-modal search
Multimodal RAG ingestion
Image and video similarity at scale
Cost-sensitive multimodal pipelines
Visual recommendation features

Quick start

cURL

curl https://aiapi-pro.com/v1/embeddings \
  -H "Authorization: Bearer $NOVAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kinfra-vl-embedding-2b",
    "input": [{"type":"text","text":"hello world"}]
  }'

Python (OpenAI SDK)

from openai import OpenAI
client = OpenAI(
    base_url="https://aiapi-pro.com/v1",
    api_key="YOUR_NOVAI_API_KEY",
)
resp = client.embeddings.create(
    model="kinfra-vl-embedding-2b",
    input=[{"type":"text","text":"hello world"}],
)
print(len(resp.data[0].embedding), "dimensions")